Hodgson Murray
School of Occupational and Environmental Hygiene and Department of Mechanical Engineering, University of British Columbia, Vancouver, Canada.
J Acoust Soc Am. 2002 Aug;112(2):568-75. doi: 10.1121/1.1490363.
Nonoptimal classroom acoustical conditions directly affect speech perception and, thus, learning by students. Moreover, they may lead to voice problems for the instructor, who is forced to raise his/her voice when lecturing to compensate for poor acoustical conditions. The project applied previously developed simplified methods to predict speech intelligibility in occupied classrooms from measurements in unoccupied and occupied university classrooms. The methods were used to predict the speech intelligibility at various positions in 279 University of British Columbia (UBC) classrooms, when 70% occupied, and for four instructor voice levels. Classrooms were classified and rank ordered by acoustical quality, as determined by the room-average speech intelligibility. This information was used by UBC to prioritize classrooms for renovation. Here, the statistical results are reported to illustrate the range of acoustical qualities found at a typical university. Moreover, the variations of quality with relevant classroom acoustical parameters were studied to better understand the results. In particular, the factors leading to the best and worst conditions were studied. It was found that 81% of the 279 classrooms have "good," "very good," or "excellent" acoustical quality with a "typical" (average-male) instructor. However, 50 (18%) of the classrooms had "fair" or "poor" quality, and two had "bad" quality, due to high ventilation-noise levels. Most rooms were "very good" or "excellent" at the front, and "good" or "very good" at the back. Speech quality varied strongly with the instructor voice level. In the worst case considered, with a quiet female instructor, most of the classrooms were "bad" or "poor." Quality also varies with occupancy, with decreased occupancy resulting in decreased quality. The research showed that a new classroom acoustical design and renovation should focus on limiting background noise. They should promote high instructor speech levels at the back of the classrooms. This involves, in part, limiting the amount of sound absorption that is introduced into classrooms to control reverberation. Speech quality is not very sensitive to changes in reverberation, so controlling it for its own sake should not be a design priority.
非理想的课堂声学条件会直接影响语音感知，进而影响学生的学习。此外，这些条件可能会给教师带来嗓音问题，因为教师在讲课时被迫提高音量以弥补声学条件不佳的状况。该项目应用先前开发的简化方法，根据在无人和有人的大学教室中的测量结果来预测在有人的教室中的语音清晰度。这些方法被用于预测不列颠哥伦比亚大学（UBC）279间教室在70%满员时不同位置的语音清晰度，以及四种教师嗓音水平下的情况。根据房间平均语音清晰度确定的声学质量，对教室进行分类并排序。UBC利用这些信息对教室翻新的优先级进行排序。在此，报告统计结果以说明一所典型大学中声学质量的范围。此外，研究了质量与相关课堂声学参数的变化，以更好地理解结果。特别是，研究了导致最佳和最差条件的因素。结果发现，在有“典型”（普通男性）教师的情况下，279间教室中有81%的声学质量为“良好”、“非常好”或“优秀”。然而，有50间（18%）教室的质量为“一般”或“较差”，还有两间质量为“差”，原因是通风噪音水平较高。大多数房间在前面是“非常好”或“优秀”，在后面是“良好”或“非常好”。语音质量随教师嗓音水平变化很大。在考虑的最差情况下，即教师是安静的女性时，大多数教室的质量为“差”或“较差”。质量也随占用情况而变化，占用率降低会导致质量下降。研究表明，新的课堂声学设计和翻新应侧重于限制背景噪音。它们应促使教室后排的教师嗓音水平较高。这部分涉及限制引入教室的吸声量以控制混响。语音质量对混响变化不太敏感，因此仅为控制混响而进行控制不应成为设计的首要任务。
